It seems vision functionality is being expanded on in the new release of gpt-4-turbo (gpt-4-turbo-2024-04-09) per openAi’s latest newsletter I received today. I have been waiting to be able to send images directly to my gpt-4-turbo assistant via assistants API for vision processing. On the assistants api docs under messages it still states: “At the moment, user-created Messages cannot contain image files but we plan to add support for this in the future.”
Although the It seems the docs aren’t showing its possible has anyone found a way to send images straight to the assistant?I have a workaround that uses completions to create text descriptions of images that are then fed back into assistants but it would be much better not having to do that any longer. I am in the process of testing new things.
Thanks all in advance